ASH (Active Session History) is the OceanBase database's "intelligent monitoring system," taking a state snapshot of active sessions every second. Through three zero-prerequisite experiments, this article lets you experience the charm of ASH and takes a deep dive into its implementation — selective recording, time slicing, and the ring buffer.
Dmall shares its hands-on experience migrating from MySQL to OceanBase in a retail SaaS scenario — using multi-tenant resource isolation, transparent scaling, and a high compression ratio to cut storage costs by over 80%, achieving cost reduction, efficiency gains, and low-cost system upgrades.
From the design of kernel components — plan management, concurrency control, multi-dimensional caching, hot-row optimization, log aggregation, and compilation optimization — this article explains the key factors behind how OceanBase safeguards performance in high-concurrency scenarios.
A practical guide to building a personalized recommendation system for UGC communities on OceanBase's native vector capabilities: dual-vector user interest modeling (short-term + long-term), multi-path recall and fusion ranking in a single SQL statement, with schema design, performance tuning, and a two-week MVP roadmap.
This article traces the four leaps of code search — from text matching to semantic retrieval, graph indexing, and agentic search — and shares hands-on experience building the CodeRepoIndex code-indexing tool on OceanBase vectorized indexes and private RAG.
CR Vanguard's subsidiary Wanjia Digital upgraded a middleware-based MySQL sharded cluster to OceanBase, achieving a seamless, business-transparent migration — system performance up 70%, storage footprint down about 60%, and hardware costs cut by 50%.
This article explains the principles and usage of hybrid queries on OceanBase vector indexes, including how to choose among the three scalar-filtering methods — Pre-filtering, In-filtering, and Iterative-Ann — as well as fusion ranking methods for full-text and vector results, and the product roadmap.
OceanBase and ModelScope jointly introduce a new paradigm for AI-driven intelligent database management. Through MCP, developers can create, run, manage, diagnose, and analyze OceanBase clusters simply by chatting in natural language.
A 99Bill DBA shares the three transformations of their DB architecture — from MySQL primary-standby and MyCAT sharding to OceanBase — detailing five migration lessons that ultimately cut hardware costs by 52%.
This article shares the practice of choosing Dify on OceanBase to build AI applications, using OceanBase to consolidate PostgreSQL, Redis, and Weaviate into a single store, and introduces real-world experience from scenarios such as psychological assessment and intelligent customer service.